Cryptococcal meningitis (CM) is a leading cause of death among all AIDS-related opportunistic infections, causing an estimated 1 million infections and 625,000 deaths worldwide annually. Even with availability of antiretroviral therapy (ART), the 6-month survival after CM is poor, with survival being only 30-40% in resource-limited areas. In two African cohorts, 35-40% mortality occurred during the interval before initiating outpatient ART at a median of 5-6 weeks after CM diagnosis. The current standard practice throughout much of the world is to delay initiation of ART until after hospital discharge to allow the establishment of outpatient care with HIV ART counseling. However, this interval prolongs the duration of immunosuppression and delays the time until ART-associated immune reconstitution occurs. Existing data about the optimal timing of ART in persons with CM are limited and conflicted. Thus, therapeutic equipoise for a randomized trial exists. Specific Aim #1: Conduct a randomized trial among persons with cryptococcal meningitis, to determine whether 26-week survival is improved with early ART initiation compared with standard ART initiation. Hypothesis: After CM, early initiation of ART during initial hospitalization during the second week of amphotericin therapy will improve survival as compared to the current standard of care. Design: After 7-10 days of amphotericin (study entry), subjects will be randomized in 1:1 allocation to: - Early ART Initiation (Experimental Group) = ART initiated within 72 hours after study entry OR - Standard ART Initiation (Control Group) = ART at >4 weeks after study entry Specific Aim #2: Determine the safety and tolerability of early ART versus standard ART with respect to the: incidence of IRIS, virologic suppression, microbiologic clearance, ART discontinuation, and adverse events. Specific Aim #3: Determine risk factors for IRIS and/or overall ART-related mortality with respect to the timing of ART after cryptococcal meningitis. The trial will enroll 420 participants to detect a 15% absolute (25% relative) difference in 26-week survival. RELEVANCE: In Sub-Saharan Africa, cryptococcal meningitis is the second most common AIDS-defining illness, and causes ~30%) of the AIDS-related attributable mortality. An estimated 1 million cases of cryptococcosis occur worldwide annually. The optimal time to start HIV therapy is unknown. This is a definitive strategy trial to determine if early initiation of HIV therapy while hospitalized results in superior long term survival.